Original articles

Spontaneous reanastomosis between lymphatic vessels following syngeneic

transplantation of the small intestine in the rat

A. Üner 3, A.-M. Weinberg4 , C. Poulsen Nautrup 5 , I. Kassianoff1 , W. Lüdemann6 , F. Schier 2 , P.

Claus1 and D. Berens v. Rautenfeld1

1 Medical School Hannover, Center of Anatomy, Carl -Neuberg-Stra ße 1, D-30625 Hannover, Germany
2 University Medical Center, Department of Pediatrics, Bachstra ße 18, D-07740 Jena, Germany
3 UCLA, Center for Prehospital Care, 924 Westwood Blvd., Suite 720, Box 951369, Los Angeles, California 90095-
1369, USA
4 Medical School Hannover, Traumatology, Carl-Neuberg-Straße 1, D-30625 Hannover, Germany
5 Ludwig-Maximilian-University, Veterinary Faculty, Institute of Veterinary Anatomy I, Veterinärstr. 13, D-80539
München, Germany
6 Medical School Hannover, Neurosurgical Department, Carl-Neuberg-Stra ße 1, D -30625 Hannover, Germany

Received October 10, 2001 / Accepted in final form August 3, 2001

Key words: Microradiology - Scanning electron microscopy - Spontaneous lymphvascular

reanastomosis - Transplantation of the small intestine - Rat

Correspondence to: Prof. Dr. D. Berens v. Rautenfeld E-Mail : Rautenfeld.Dirk@mh -hannover.de

Abstract

Abstract Spontaneous lymphvascular reanastomosis (SLR) following small bowel transplantation in

rats is of clinical relevance for the resorption of long chain fatty acids. Detailed morphological and
molecular data concerning the process of lymphvascular reanastomosis are not available in the
literature. In this study SLR was investigated using microradiology and scanning electron
microscopy. Between the 8th and 21st postoperative days following transplantation SLR does not
occur between the intestinal trunk of the transplant and the thoracic duct of the recipient. Instead, an
indirect connection was observed between the inserted advential lymphatic vessels of the mesenteric
artery and lymphatic vessels of the aorta or ductus deferens, which are connected with the thoracic
duct.

Surgical reanastomosis of interrupted lymphatic vessels in organ transplantations is not performed in

man, hence temporary postoperative edema of the graft may be anticipated. When performing
transplantations of the small intestine, the efficacy of spontaneous or surgical lymphvascular
reanastomosis must be critically assessed because of the absorption of fat through the lymph vessels.
Spontaneous lymphvascular reanastomosis (SLR) of the graft is of significance after transplantation
of the small intestine for short bowel syndrome, as is intestinal lymphangiectasis [6]. In the present
study "SLR" is defined as the spontaneous (non-surgical) reanastomosis of two lymphatic vessels
between transplant and recipient.

Following transplantation of the small intestine in the rat, it is possible to perform surgical
reanastomosis [10] of the lymphatic drainage ducts of the transplant with the thoracic duct [11].
There are controversial animal experimental data concerning the time span of the corresponding
spontaneous lymphvascular reanastomosis 3 days for the rat [18] and 4 weeks for the pig [17].

Conventional radiological and microscopic investigative techniques have been used unsuccessfully
in providing a morphological characterization of the progressive lymphvascular reanastomosis
between transplant and recipient.

In the present study microradiological and scanning electron microscopic investigations were
conducted to investigate possible problems with lymphatic drainage after organ transplantation.
Using this combined investigative procedure, devised by Poulsen Nautrup and Berens v. Rautenfeld
[14], it was possible to demonstrate both the radiological and microstructural aspects of the initial
lymph vessel reanastomosis.

Material and methods

Syngeneic heterotopic transplantation of the small intestine was performed on 22 male Lewis rats
using the technique of Preissner et al [15]. The weight of the donors was on average 311 g and that
of the recipients 325 g.

Transplantation technique

The small intestine was removed from the donor from the lig. of Treitz to 2 cm proximal to the
ileocaecal valve with the mesentery, the cranial mesenteric v., the portal v., the cranial mesenteric a.
with short aortic segments and the jejunal lymph nodes with the intestinal trunk. The transplants
were stored in 0.9% sodium chloride solution until the time of transplantation. The average ischemic
timewas 61 ± 10min.

The aorta and caudal vena cava of the recipient were surgically prepared approximately 1 cm distal
to the renal v. and clamped proximally and distally. Using an operating microscope end-to-side
anastomoses were performed between the portal v. of the transplant and the caudal vena cava of the
recipient, as well as between the cranial mesenteric a. via the aortic segment of the transplant and the
aorta (Fig. 1) of the recipient. The transplanted and separated intestinal trunk was left exposed and
was not reanastomosed. The proximal jejunal end of the transplant was closed as a blind pouch,
while the ileal end was anastomosed end-to-side at the exposed antimesenteric ileum of the recipient
close to the ileocaecal valve.
Fig. 1 Schematic representation of the spontaneous lymphvascular reanastomosis following
transplantation of the small intestine. The testicular lymphatic vessels flow into the thoracic duct
cranial to the mesenteric a.. 1, aorta 2, cranial mesenteric a. 3, area of the arterial end-to-side
anastomosis 4, intestinal trunk 5, initial adventitial lymphatic vessels 6, left testicular lymphatic
vessels associated to the vein and artery 7, thoracic duct

Microlymphography

The recipients of the transplant (n = 22) were subdivided into 6 groups. The lymphographic and
scanning electron microscopic investigations were performed on the 2 nd (n = 3), 4 th (n = 4), 6 th
(n = 4), 8 th (n = 4), 10 th (n = 4) and 21 st (n = 3) postoperative days, respectively.
Microlymphography was carried out using a protective X-ray system FXS 100.22 (Feinfocus X-ray
System Co., Wunstorf).

The rats were first anaesthetized with Ketanest® and Rompun® and laparotomized in the supine
position. The small intestine transplant was laid free of pressure and placed to the left side of the
animal. Each rat received X-ray contrast medium Iotasul ® (developing preparation) or Isovits® 300
(Schering Co., Berlin) using an injector pump (Precidor 5003, Heinemann Co., Schwäbisch Gmünd)
for a period of 90mn with an administration rate of 0.02175 ml/mn into one of the jejunal lymph
nodes of the transplant. Under identical perfusion conditions X-ray contrast medium was placed into
the interstitium of the left testicle in 8 of the 22 rats to fill the testicular lymphatic vessels. For this
procedure the rat was placed on a Plexiglas plate together with the injector within the closed X-ray
system.

Microlymphographic documentation was undertaken using still radiographs with a high-resolution

film (AGFA Structurix D7 size 18 x 24 cm), as well as continuous fluoroscopic videos beginning
5mn and continuing to 90mn into the perfusion of each animal.

To increase the quality of the video film, subtraction angiography or pseudo-3D-presentation was
introduced using deviation-image subtraction with an image evaluation system Hamamatsu DVS
3000.
Scanning electron microscopy (SEM) procedure

Intranodal perfusion of Iotasul ® and Isovist ® was also conducted for rinsing and distending the
operative site in preparation for perfusion fixation of the lymph drainage tract between transplant and
recipient. Immediately following lymphvascular rinsing (after 90 minutes), intranodal perfusion
fixation was performed with 2.5% glutaraldehyde using the same administration rate
(0.02175 ml/mn) for up to 180mn.

From each of the 26 animals, 5 SEM specimens were taken from the following areas perfused jejunal
lymph nodes, intestinal trunk, mesentery, cranial mesenteric a. and aorta. Preparation and assessment
of the SEM specimens were performed according to the method of Berens v. Rautenfeld et al. [5]
using the Philips SEM 505.

Results

Spontaneous lymphvascular reanastomosis (SLR) took place in two chronological phases, as shown
by microfocus lymphography and scanning electron microscopy (SEM).

Phase I Provisional lymphvascular reanastomosis (2nd to 6th postoperative days)

On the 2nd postoperative day, after application of Isovist ® into one of the jejunal lymph nodes,
abrupt termination of contrast medium in the region of the intestinal trunk of the transplant resulted
in all investigated animals (Fig. 2). Only one of three animals showed a contrast medium flow
through the extranodular veins of the jejunal lymph nodes into the caudal vena cava of the recipient
on the 2 nd postoperative day. SEM showed evidence of intranodular, as well as dilated lymph sinuses
and veins, however no lymphovenous anastomoses were detected. Eighteen of the 22 animals
showed extranodal venous contrast medium flow in each group from the 6th to the 21st postoperative
day. No evidence of other lymph vessels within and outside the transplant was found.

Fig. 2 Microfocus lymphography after application of Isovist® into one of the jejunal lymph nodes,
2nd postoperative day (60kV 0,2mA 35sec direct radiological magnification = 4x). 1, injected jejunal
lymph node 2 intestinal trunk with contrast medium
On the 4th postoperative day, in addition to termination of the contrast medium in the intestinal
trunk, a network of initial lymphatic vessels was found close to the cranial mesenteric arterial wall
within the transplant (Fig. 1). Contrast medium was seen as far as the anastomosis of the cranial
mesenteric a. of the transplant with the aorta of the recipient, however no contrast was seen to cross
into the recipient. Using SEM the initial adventitial lymphatic vessels demonstrated typical structural
characteristics of the lymphvascular endothelial layer interendothelial openings, endothelial bridges,
connective tissue trabeculae and subendothelial filaments. Within the intestinal trunk distended pore-
like endothelial openings were seen in the form of outflow valves. It was not possible to detect any
lymphvascular or interstitial channels containing contrast medium between the intestinal trunk and
the adventitial network of the initial lymphatic vessels (lymphatic capillaries and precollectors).

On the 6th postoperative day SLR was observed in one animal between the adventitial initial
lymphatic vessels in the wall of the cranial mesenteric a. of the transplant and the adventitial initial
lymphatic vessels of the aorta of the recipient. This was confirmed by SEM (Fig. 3).

Fig. 3 SEM wall of the cranial mesenteric a. near to the aorta. 1, intravasal lumen 2, intima 3, media
4, suture material (Ethicon ®) 5, adventitia 6, initial lymphatic vessels (capillaries and precollectors)

Phase 2 Final lymphvascular reanastomosis (from the 8 th postoperative day)

All four animals studied on the 8 th, and all animals studied up to the 21st postoperative days
demonstrated SLR in the adventitia of the mesenteric a. and the aorta and a new lymphvascular
reanastomosis (Figs. 4, 5). This new connection was formed between one or two markedly extended
lymphatic vessels in the adventitia of the mesenteric a. of the graft and the ascending testicular
lymphatic vessels of the recipient. Renal and lumbar paraaortic lymph nodes can drain into this
lymphatic channel.
Fig. 4 Microfocuslymphography after application of Isovist® into one of the jejunal lymph nodes and
into the left testis on the 8th postoperative day (70kV, 0,1mA, 80sec, direct radiographic
magnification = 4x). 1, injected jejunal lymph node 2, cranial mesenteric a. with initial adventitial
lymphatic vessels (capillaries and precollectors) 3, initial adventitial lymphatic vessels in the area of
the arterial anastomosis between the cranial mesenteric a. and the aorta 4, left testicular lymphatic
vessel 5, left renal lymph node 6, thoracic duct
Fig. 5 Microfocuslymphography after application of Isovist® into one of the jejunal lymph nodes and
into the left testis on the 21st postoperative day (60kV, 0,2mA, 35sec, direct radiographic
magnification = 4x). 1, injected jejunal lymph node 2, cranial mesenteric a. with initial adventitial
lymphatic vessels (capillaries and precollectors) 3, lymphatic vessel (collector) 4, testicular
lymphatic vessels 5, thoracic duct 6, ureter 7, kidney

Discussion

The study has demonstrated that the initial lymphatic vessels form a reparative outflow passage
between the transplant and donor. Radiological evidence of the initial lymphatic vessels can only be
determined microlymphographically [14], which with electron microscopic investigations provide
supplementary evidence of the typical structural elements of the initial lymphatic vessels [4].
Bellmann and Oden [3] and Walzer [22] were unable to characterize ‘lymphatic vessel
regeneration' as initial lymphatic vessels after separation of lymphatic vessels in rabbit ears.

The first evidence of contrast medium outflow from the lymph system into the blood system in the
current series was seen in the contrasted jejunal lymph nodes of the graft. This occurred in one
animal on the 2nd postoperative day and subsequently in 18 of the animals. This lymphovenous
outflow could indicate a temporary overflow valve function prior to regeneration of lymphovenous
anastomoses. Intranodal lymphovenous anastomoses can appear even after experimental sealing of
the thoracic duct [8]. In contrast to humans, extranodal lymphovenous anastomoses can be regularly
observed in animals [2]. However, no physiological lymphovenous anastomoses have been found
within the lymph nodes of the rat [7], indicating that this outflow phenomenon could be an artificial
lymphovenous anastomosis.

As expected SLR did not take place directly between the intestinal trunk of the transplant and the
thoracic duct of the recipient. Instead, from the 8 th postoperative day initial lymphatic vessels
provided a lymphvascular outflow passage from the wall of the cranial mesenteric a. of the transplant
to the adventitial lymphatics of the aorta of the recipient. The arterial anastomosis between the
mesenteric a. of the graft and the aorta of the recipient approximates the adventitia of both vessels,
enabling the first by-way anastomosis of the adventitial initial lymphatic vessels of both arteries
(Fig. 3).

The final phase of SLR was demonstrated in all animals after the 8 th postoperative day (Figs. 4, 5),
being characterized by significantly dilated initial lymphatic vessels in the mesenteric arterial wall,
draining via the testicular lymph vessels to the thoracic duct [21]. Schmid et al. [18] have reported a
primary lymphvascular anastomosis following transplantation of the small intestine in the rat on the
3rd postoperative day, however without any anatomic characterization of the outflow passage.

Two competing models for the new formation of lymphatic vessels are considered sprouting of
lymphatic vessels from the venous system (centrifugal or venous theory of Ranvier [16]) and the
formation of isolated interstitial, and subsequently endothelialized, vessels from the mesenchyme
(central or mesenchymal theory of Gulland [9]). However, microradiography does not allow
discrimination between these two models, since it is not possible to detect isolated lymphatic vessels.
In subsequent studies reconstructions of serial sections, in combination with molecular markers may
be helpful. To determine the underlying molecular processes for the observed new formation of the
initial lymph vessels between the intestinal trunk of the graft and the adventitia of its cranial
mesenteric a. in this study, molecular biological methods need to be applied. It has recently been
shown that vascular endothelial growth factors -C and -D (VEGF-C, -D), respectively and the
VEGF-receptor-3 are key molecules for lymphangiogenesis [13, 23]. Their expression correlates
with lymphatic metastasis in mouse tumor models [12, 19, 20]. In addition, lymphatic vessel
endothelial hyaluronan receptor -1 (LYVE-1), a homologue of the glycoprotein CD44, is expressed
almost exclusively in lymphatic vessels, but is absent in blood vessels [1]. These molecules will be
used in subsequent investigations of the intestine transplantation model as molecular markers and for
exogenous applications.

References

1. Banerji S, Ni J, Wang SX, Clasper S, Su J, Tammi R, Jones M, Jackson DG (1999) LYVE -1, a
new homologue of the CD44 glycoprotein, is a lymph-specific receptor for hyaluronan. J Cell Biol
144 789-801

2. Belz L. (1980) Lymphovenous Anastomoses. In Viamonte M. Rüttimann Atlas of Lymphography.

Thieme-Stratton, New York, pp 187-188

3. Bellmann S, Oden B (1959) Regeneration of surgical divided lymph vessels. An experimental

study on the rabbit´s ear. Acta Chir Scand 116 99-117

4. Berens v. Rautenfeld D, Lubach D, Wenzel-Hora BI, Buchholz T, Poulsen Nautrup C (1987) New
techniques of demonstrating lymph vessels in skin biopsies and intact skin with the scanning electron
microscope. Arch Dermatol Res 279 327-334

5. Berens v. Rautenfeld D, Wenzel-Hora BI, Lubach D, Hunneshagen C, Deutsch A (1989) Vacuum

injection - A new indirect fixing method for the demonstration of initial lymphatics in biopsy
specimens of human organs. Acta Anat 136 172-176

6. Booth CC, Neale G (1985) Disorders of the Small Intestine. Blackwell Scientific, Oxford, pp 348-
359

7. Engeset A (1959) Lymphatico-venous communications in the albino rat. J Anat 92 380-383

8. Hidden G., Menard Ph, Zorn JY (1985) Lymphaticovenous communications. Anat Clin 7 83-91
9. Gulland GL (1884) The development of lymphatic glands. J Pathol Bact 2 447-485

10. Kellersmann R, Grant D, Zhong R (1999) Technical experiences with a microsurgical model of
lymphatic reconstruction after intestinal transplantation in rats. Microsurgery 19 108-112

11. Kressel M, Baumeister RGH, Gabka CH (1990) Erste lympho-lymphatische Anastomisierung bei
der Dünndarmtransplantation der Ratte. Lymphologica Jahresband. Medikon, München, pp 103-105

12. Mandriota SJ, Jussila L, Jeltsch M, Compagni A, Baetens D, Prevo R, Banerji S, Huarte J,
Montesano R, Jackson DG, Orci L, Alitalo K, Christofori G, Pepper MS (2001) Vascular endothelial
growth factor-C-mediated lymphangiogenesis promotes tumour metastasis. EMBO J 20 672-682

13. Oh SJ, Jeltsch MM, Birkenhäger R, McCarthy JEG, Weich HA, Christ B, Alitalo K, Wilting J
(1997) VEGF and VEGF-C Specific induction of chorioallantoic membrane. Dev Biol 188 96-109

14. Poulsen Nautrup C, Berens v. Rautenfeld D (1991) Direktradiographische Vergrößerung in der

experimentellen Medizin. Radiologe 31 430-434

15. Preissner WC, Schroeder P, Gundlach M, Liedgens P (1986) Microsurgical Techniques of

heterotopic and orthotopic small-bowel transplantation in the rat. In Small-Bowel Transplantation,
Springer, pp 2-6

16. Ranvier L (1895) Développement des vaisseaux lymphatiques. C R Acad Sci 121 1105-1109

17. Rotman N, Michot F, Hay JM, Fragniez PL (1986) Lymphatic regeneration following intestinal
transplantation in the pig. Small-Bowel Transplantation. Springer, Berlin, pp 34-36

18. Schmid T, Körösci G, Klima G, Margreiter R (1988) Untersuchungen zum Lymphabfluß aus
Dünndarmtransplantaten (DTXS) bei Ratten. Acta Chir Austr 3 105-106

19. Skobe M, Hawighorst T, Jackson DG, Prevo R, Janes L, Velasco P, Riccardi L, Alitalo K,
Claffey K, Detmar M (2001) Induction of tumor lymphangiogenesis by VEGF -C promotes breast
cancer metastasis. Nature Med 7 192-198

20. Stacker SA, Caesar C, Baldwin ME, Thornton GE, Williams RA, Prevo R, Jackson DG,
Nishikawa S, Kubo H, Achen MG (2001) VEGF-D promotes the metastatic spread of tumor cells via
the lymphatics. Nature Med 7 186-191

21. Üner A, Berens v. Rautenfeld D, Poulsen Nautrup C, Schier F, Wenzel-Hora BI, Schroeder P,
Waldschmidt J (1990) Mikrolymphographie der spontanen Lymphgefäß - Reanastomisierung nach
syngener Dünndarmtransplantation bei der Ratte. Lymphologica Jahresband, Medikon, München, pp
98-102

22. Walzer LR (1986) Zur Regeneration des Lymphgefäßsystems nach Replantation und
Insellappenplastik. Acta Chir Austr Suppl 66, Brüder Hollinek, Wien

23. Wilting J, Neeff H, Christ B (1999) Embryonic lymphangiogenesis. Cell Tissue Res 297 1-11

Surg Radiol Anat 23: 383-387

© Springer-Verlag France 2001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.